Feb., 1937
THECONFIGURATIONS OF a- AND P-~-BROMOBENZOPHENONE OXIME
365
a-d-Mannose Pentaacetate.-Pure 8-d-mannose pentaacetate (3.0 9.) was treated exactly as described for 6 4 glucose pentaacetate except that the time of shaking wa.. from four to four and one-half hours. The acidified filtrate showed a specific rotation of approximately 4-45". The sirup obtained after solvent removal was crystallized from hot water; yield 1.2 g.; m. p. 73-74"; [alas +59" ( 6 . 2.6; CHCII; 2-dm. semimicro tube). After one further recrystallization the substance showed the constants: m. p. 74-75' (mixed m. p. unchanged); [a]% +60' ( c , 3.3; CHCb; 2-dm. semimicro tube). Levene and BencowitzQ record for a-d-mannose pentaacetate the constants: m. p. 75"; [CY]'* +55' (CHCla). a-Lactose 0ctaacetate.-Pure 0-lactose octaacetate (3.0 g.; m.p. 90'; [a]-4.2", CHCla) was treated exactly as described for 8-d-glucose pentaacetate and the sirup obtained after solvent removal was dissolved in ethanol (Carboraffin) and the ethanol removed under reduced pressure The residual sirup was triturated with several portions of cold water and obtained crystalline from either ethanol or ether by slow evaporation of the solvent. aLactose octaacetate is difficult to crystallize and the yield was very low; m. p. 149-150"; [a]2' +51° (c, 2.9; CHCb; 2-dm. semimicro tube). On one further recrystallization the melting point was 152' (mixed m. p. unchanged). Hudson and Johnsonlb record for a-lactose octaacetate the constants: ni. p. 152'; [a]"f54" (CHCls).
This point was settled satisfactorily when it was found that ether alone could be used as a solvent if the time of shaking was prolonged to from eighteen to twenty-four hours. Purified dioxane could also be used alone but then the reaction likewise proceeded a t a very slow rate. The sodium used in the procedure served merely as a drying agent, since anhydrous calcium sulfate (Drierite) could be substituted for it without affecting the yield. When sodium was used and the solid sodium hydroxide omitted, a very low yield of a-glucose pentaacetate was obtained. This small amount of conversion was undoubtedly due to the slight quantity of sodium hydroxide produced by unavoidable contamination with moisture. All of these experiments pointed to the conclusion that the active interconverting agent was the sodium hydroxide. When a-d-glucose pentaacetate was used initially, it was recovered unchanged in approximately 50Yo yield but no &form was isolated. This would indicate either that the equilibrium was displaced to favor the a-form or that the @-isomerwas more readily deacetylated. a-d-Galactose Pentaacetate.-Pure B-d-galactose pentaacetate (3.0 9.) was treated exactly as described for @-dglucose pentaacetate except that the time of shaking was from five to seven hours. The acidified filtrate showed a specific rotation of from $60 to $80". The sirup obtained after solvent removal under reduced pressure was dissolved in ethanol, treated with Carboraffin, filtered and the ethanol removed under reduced pressure. The residual sirup was dissolved in chloroform and the solution washed successively with 5% hydrochloric acid, 5% aqueous sodium bicarbonate and then with water, dried, the solvent removed under reduced pressure and the residual sirup crystallized from a small amount of hot ethanol; yield 0.8 to 1.0 g.; m. p. 94-95' (mixed m. p. unchanged); CY]^' $104' (c, 3.7; CHClS). Hudson and Parkefl record for a-d-galactose pentaacetate the constants: m. p. 96 O; [a]" $107' (CHCb).
I . A method was found for converting P-dglucose pentaacetate to the a-form, using sodium hydroxide as the reagent. 2 . The method was extended to the corresponding fully acetylated derivatives of d-galactose, d-mannose and lactose. It is probably of general application.
(8) C. S. Hudson and H. 0. Parker, THISJOURNAL, 37, 1689 (1816).
COLUMBUS, OHIO
[CONTRIBUTION FROM
THE
Summary
(9) P. A. Levene and I . Bencowitz, J. B i d .
Chem., 79, 627 (1927)
RECEIVED NOVEMBER19, 1936
GEORGEHERBERT JONKS CHEMICAL LABORATORY OF THE UNIVERSITY OF CHICAGO]
The Configurations of ~ 1 - and p-p-Bromobenzophenone Oximes BY RAYMOND WRIGHTJOHNSON An earlier contribution by Johnson and Stieglitz' from this Laboratory reported the details of an empirical method for the determination of the velocity of hydrolysis of oximes. Data were reported which showed that the individual members of the pair of stereoisomeric oximes of pmethylbenzophenone hydrolyzed at considerably different rates. The purpose of this study was, first, to test the generality of this difference in rates by the investigation of another pair of stereoisomeric oximes and, second, to extend the study to include a com(1) Johnson and Stieglitx, THISJOURNAL, 116,1904 (1934).
AND JOE
GOLENTERNEK
parison of the rates of hydrolysis of the stereoisomeric forms with those of the corresponding symmetrical oximes. It was hoped that this might throw some light on the configurations of the stereoisomeric compounds. The compounds selected were the a and p forms of p-bromobenzophenone oxime and the related oximes of benzophenone and #,fir-dibromobenzophenone. Preparation of Materials Benzophenone Oxime.-This compound was prepared by the method of Fischer.* (2) Pischer, "Anleitung zur Darstellung organ. PrepSrate." Braunxhweig, 1908, p. 68.
366
RAYMOND WRIGHTJOHNSON
p,p'-Dibromobenzophenone Oxime.-A mixture of the ketone (10 g.) and hydroxylamine hydrochloride (6.5 g.) dissolved in ethyl alcohol (700 cc.of 95%) to which a solution of sodium carbonate (10 g. in 30 cc. of water) had been added was refluxed on a steam-bath for ten hours and then allowed to cool. The unreacted ketone (about 26% of the original amount) and sodium carbonate was removed by filtration. The filtrate was diluted with water (2500 cc.) and chilled. Crystals of the crude oxime (7.5 g., m. p. 14G145") were obtained. Two recrystallizations From ligroin (b. p. 90-100') produced the pure oxime(m. p. 160-162 "). Anal. (Micro) Calcd. for Cl&ONBrs: N, 3.94. Found: N, 3.94, 4.02. a- and 8-PBromobenzophenone Oximes.-Anhydrous sodium acetate (100 g.) was added to a solution of. the ketone (50 g.) and hydroxylamine hydrochloride (60 g.) in ethyl alcohol (400 cc. of 95%). The mixture was refluxed for six hours on a steam-bath. The alcohol was removed by distillation and the residue was washed several times with warm water. The crude mixture of the two oximes (51 g . ) when dry had a melting point of 102-160". The method used for the separation of the two forms was based on the difference m their solubility in solutions of sodium hydroxide. The crude mixture of oximes was extracted with successive portions (700, 150, 85 and 50 cc.) of a hot (90") solution of sodium hydroxide (2 N). The undissolved oxime (m. p. 160-165') on recrystallization from ethyl alcohol produced the pure a-oxime. The melting point of this pure a-form (m. p. 168-170") was higher than that reported (166-168') previously by Schiifer.8 The combined alkaline extract was saturated with carbon dioxide gas to precipitate the oxime in solution. This material was washed with water until free from alkali. The dried material was dissolved in hot ethyl alcohol (absolute). Fractional crystallization was obtained by means of the addition of small quantities of water from time to time. Relatively small amounts of the pure p-form (m. p. 109111') were obtained. Anal. (Micro) Calcd. for CI3H100NBr: N, 5.07. Found: (a-form), N , 5.01,4.93; (p-form), 5.03,4.85. Method The method used for the determination of the veiocity of hydrolysis of the oximes was essentially that of Johnson and Stieg1itz.l The limited solubility of these bromo compounds in an acetic-hydrochloric acid mixture such as was used by them required the use of a mixture richer in acetic acid. This necessitated a slight change in the extraction procedure. The procedure followed in each of the determinations listed below was identical. A weighed quantity of the oxime was dissolved in glacial acetic acid (140 cc. of reagent quality) a t 25" in a 200-cc. graduated flask. At a definite noted time exactly 50 cc. of hydrochloric acid (1.899 N ) a t %' was rtlfl in. The volume of t h e solution was quickly made up to 200 cc. by the addition of glacial acetic acid. After it was thoroughly mixed, the solution was placed in a t h e m s t a t a t 25 * 0.02". From time to time a sample (25 cc.) of the solution was pipetted into a Squibb separatory funnel. At a definite noted time 35 cc. of water was added and the proceduie ( 3 ) SchBfer, A n n . , 264, 162 (1801).
AND JOE
GOLENTERNEK
Vol. 59
from that point on was the Sanie as that described by Johnson and Stieglitz.1 The concentration of the oxime in the various determinations ranged from 0.03 to 0.045 N Tests with known solutions showed that this procedure yields quantitative results with acetic-hydrochloric acid solutions of hydroxylamine such as were obtained in this investigation. The fact that the reactions in each case were normal and yielded only the respective ketone and hydroxylamine was proved when examinations of the content of completely hydrolyzed solutions yielded only these two substances. Results The average of a t least two determinations of the velocity of hydrolysis of each of the oximes studied is given in Table I. TABLE I VELOCITY OF HYDROLYSIS OF OXIMhb (25") Compound, oxime
Benzophenone a-#-Bromobenzophenone 8-p-Bromobenzophenone p,p'-Dibromobenzophenone
Concn. of HCI in hydrolyzing solution, N
0.4747 4747 ,4747 4747
I
